Specifications

Troubleshooting Resources 4–13
Write-Through Caching
When the controller receives a write request from the host, it places the data in its cache
module, writes the data to the disk drives, then notifies the host when the write operation is
complete. This process is called write-through caching because the data actually passes
through—and is stored in—the cache memory on its way to the disk drives.
Write-Back Caching
Write-back caching improves the subsystem response time to write requests by allowing
the controller to declare the write operation “complete” as soon as the data reaches its
cache memory. The controller performs the slower operation of writing the data to the disk
drives at a later time.
By default, write-back caching is enabled for all units. In either case, the controller will
not provide write-back caching to a unit unless the cache memory is non-volatile, as
described in the next section.
Fault-Tolerance for Write-Back Caching
The cache module supports nonvolatile memory and dynamic cache policies to protect the
availability of its unwritten (write-back) data.
Nonvolatile Memory
Except for disaster-tolerant supported mirrorsets, the controller can provide write-back
caching for storage units as long as the controller cache memory is nonvolatile. In other
words, to enable write-back caching, connect a backup power source (an ECB) to the
cache module to preserve the unwritten cache data in the event of a power failure. If cache
memory were volatile—that is, if it did not have a backup power source—the unwritten
cache data would be lost during a power failure.
By default, the controller expects to use an ECB as the backup power source for its cache
module. However, if the subsystem is backed up using a UPS, two options are available:
■ Use both the ECB and the UPS together (the NODE_ONLY setting).
■ Use only the UPS as the backup power source (the DATACENTER_WIDE setting).